When I use take(1), it will console.log twice 1, like below code:
const a$ = new BehaviorSubject(1).pipe(publishReplay(1), refCount());
a$.pipe(take(1)).subscribe();
a$.subscribe((v) => console.log(v)); // emit twice (1 1)
But when I remove take(1) or remove publishReplay(1), refCount(), it follow my expected (only one 1 console.log).
const a$ = new BehaviorSubject(1).pipe(publishReplay(1), refCount());
a$.subscribe();
a$.subscribe((v) => console.log(v)); // emit 1
// or
const a$ = new BehaviorSubject(1);
a$.pipe(take(1)).subscribe();
a$.subscribe((v) => console.log(v)); // emit 1
Why?
Version: rxjs 6.5.2
Let's first have a look at how publishReplay is defined:
const subject = new ReplaySubject<T>(bufferSize, windowTime, scheduler);
return (source: Observable<T>) => multicast(() => subject, selector!)(source) as ConnectableObservable<R>;
multicast() will return a ConnectableObservable, which is an observable that exposes the connect method. Used in conjunction with refCount, the source will be subscribed when the first subscriber registers and will automatically unsubscribe from the source when there are no more active subscribers. The multicasting behavior is achieved by placing a Subject(or any kind of subject) between the data consumers and the data producer.
() => subject implies that the same subject instance will be used every time the source will be subscribed, which is an important aspect as to why you're getting that behavior.
const src$ = (new BehaviorSubject(1)).pipe(
publishReplay(1), refCount() // 1 1
);
src$.pipe(take(1)).subscribe()
src$.subscribe(console.log)
Let's see what would be the flow of the above snippet:
src$.pipe(take(1)).subscribe()
Since it's the first subscriber, the source(the BehaviorSubject) will be subscribed. When this happens, it will emit 1, which will have to go through the ReplaySubject in use. Then, the subject will pass along that value to its subscribers(e.g take(1)). But because you're using publishReplay(1)(1 indicates the bufferSize), that value will be cached by that subject.
src$.subscribe(console.log)
The way refCount works is that it first subscribes to the Subject in use, and then to the source:
const refCounter = new RefCountSubscriber(subscriber, connectable);
// Subscribe to the subject in use
const subscription = connectable.subscribe(refCounter);
if (!refCounter.closed) {
// Subscribe to the source
(<any> refCounter).connection = connectable.connect();
}
Incidentally, here's what happens on connectable.subscribe:
_subscribe(subscriber: Subscriber<T>) {
return this.getSubject().subscribe(subscriber);
}
Since the subject is a ReplaySubject, it will send the cached values to its newly registered subscriber(hence the first 1). Then, because there were no subscribers before(due to take(1), which completes after the first emission), the source will be unsubscribed again, which should explain why you're getting the second 1.
If you'd like to get only one 1 value, you can achieve this by making sure that every time the source is subscribed, a different subject will be used:
const src$ = (new BehaviorSubject(1)).pipe(
shareReplay({ bufferSize:1, refCount: true }) // 1
);
src$.pipe(take(1)).subscribe()
src$.subscribe(console.log)
StackBlitz.
Related
When using Dependency injection in Angular I often need to subscribe to an observable that I haven't yet created!
I often end up using something like this:
// create behavior subject OF Observable<number>
const subject = new BehaviorSubject<Observable<number>>(EMPTY);
// subscribe to it, using flatMap such as to 'unwrap' the observable stream
const unwrappedSubject = subject.pipe(flatMap((x: number) => x));
unwrappedSubject.subscribe(s => console.log(s));
// now actually create the observable stream
const tim = timer(1000, 1000);
// set it into the subject
subject.next(tim);
This uses flatMap to 'unwrap' the observable contained in the subject.
This works fine, but frankly it always feels 'icky'.
What I really want is something like this, where the consumer of the subject treats the instance of the Subject as Observable<number> without having to pipe it every usage.
const subject = new UnwrappingBehaviorSubject<number>(EMPTY);
subject.subscribe((x: number) => console.log(x));
// this could use 'next', but that doesn't feel quite right
subject.setSource(timer(1000, 1000));
I'm aware that I could subscribe to the timer and hook it up directly to the subject, but I also want to avoid an explicit subscribe call because that complicates the responsibility of unsubscribing.
timer(1000, 1000).subscribe(subject);
Is there a nice way to achieve this?
The Subject.ts and BehaviorSubject.ts source files get more complicated than I expected. I'm scared I'll end up with horrible memory leaks if I try to fork it.
I think this would be another way to solve it:
foo.component.ts
export class FooComponent {
private futureObservable$ = new Observable(subscriber => {
// 'Saving' the subscriber for when the observable is ready.
this.futureObservableSubscriber = subscriber;
// The returned function will be invoked when the below mentioned subject instance
// won't have any subscribers(after it had at least one).
return () => this.futureObservableSubscription.unsubscribe();
}).pipe(
// You can mimic the Subject behavior from your initial solution with the
// help of the `share` operator. What it essentially does it to *place*
// a Subject instance here and if multiple subscriptions occur, this Subject instance
// will keep track of all of them.
// Also, when the first subscriber is registered, the observable source(the Observable constructor's callback)
// will be invoked.
share()
);
private futureObservableSubscriber = null;
// We're using a subscription so that it's easier to collect subscriptions to this observable.
// It's also easier to unsubscribe from all of them at once.
private futureObservableSubscription = new Subscription();
constructor (/* ... */) {};
ngOnInit () {
// If you're using `share`, you're safe to have multiple subscribers.
// Otherwise, the Observable's callback(i.e `subscriber => {...}`) will be called multiple times.
futureObservable$.subscribe(/* ... */);
futureObservable$.subscribe(/* ... */);
}
whenObservableReady () {
const tim = timer(1000, 1000);
// Here we're adding the subscription so that is unsubscribed when the main observable
// is unsubscribed. This part can be found in the returned function from the Observable's callback.
this.futureObservableSubscription.add(tim.subscribe(this.futureObservableSubscriber));
}
};
Indeed, a possible downside is that you'll have to explicitly subscribe, e.g in the whenObservableReady method.
With this approach you can also have different sources:
whenAnotherObservableReady () {
// If you omit this, it should mean that you will have multiple sources at the same time.
this.cleanUpCrtSubscription();
const tim2 = timer(5000, 5000);
this.futureObservableSubscription.add(tim2.subscribe(this.futureObservableSubscriber));
}
private cleanUpCrtSubscription () {
// Removing the subscription created from the current observable(`tim`).
this.futureObservableSubscription.unsubscribe();
this.futureObservableSubscription = new Subscription();
}
So I'm relatively inexperienced with rxjs so if this is something that would be a pain or really awkward to do, please tell me and I'll go a different route. So in this particular use case, I was to queue up updates to send to the server, but if there's an update "in flight" I want to only keep the latest item which will be sent when the current in flight request completes.
I am kind of at a loss of where to start honestly. It seems like this would be either a buffer type operator and/or a concat map.
Here's what I would expect to happen:
const updateQueue$ = new Subject<ISettings>()
function sendToServer (settings: ISettings): Observable {...}
...
// we should send this immediately because there's nothing in-flight
updateQueue$.next({ volume: 25 });
updateQueue$.next({ volume: 30 });
updateQueue$.next({ volume: 50 });
updateQueue$.next({ volume: 65 });
// lets assume that our our original update just completed
// I would now expect a new request to go out with `{ volume: 65 }` and the previous two to be ignored.
I think you can achieve what you want with this:
const allowNext$ = new Subject<boolean>()
const updateQueue$ = new Subject<ISettings>()
function sendToServer (settings: ISettings): Observable { ... }
updateQueue$
.pipe(
// Pass along flag to mark the first emitted value
map((value, index) => {
const isFirstValue = index === 0
return { value, isFirstValue }
}),
// Allow the first value through immediately
// Debounce the rest until subject emits
debounce(({ isFirstValue }) => isFirstValue ? of(true) : allowNext$),
// Send network request
switchMap(({ value }) => sendToServer(value)),
// Push to subject to allow next debounced value through
tap(() => allowNext$.next(true))
)
.subscribe(response => {
...
})
This is a pretty interesting question.
If you did not have the requirement of issuing the last in the queue, but simply ignoring all requests of update until the one on the fly completes, than you would simply have to use exhaustMap operator.
But the fact that you want to ignore all BUT the last request for update makes the potential solution a bit more complex.
If I understand the problem well, I would proceed as follows.
First of all I would define 2 Subjects, one that emits the values for the update operation (i.e. the one you have already defined) and one dedicated to emit only the last one in the queue if there is one.
The code would look like this
let lastUpdate: ISettings;
const _updateQueue$ = new Subject<ISettings>();
const updateQueue$ = _updateQueue$
.asObservable()
.pipe(tap(settings => (lastUpdate = settings)));
const _lastUpdate$ = new Subject<ISettings>();
const lastUpdate$ = _lastUpdate$.asObservable().pipe(
tap(() => (lastUpdate = null)),
delay(0)
);
Then I would merge the 2 Observables to obtain the stream you are looking for, like this
merge(updateQueue$, lastUpdate$)
.pipe(
exhaustMap(settings => sendToServer(settings))
)
.subscribe({
next: res => {
// do something with the response
if (lastUpdate) {
// emit only if there is a new "last one" in the queue
_lastUpdate$.next(lastUpdate);
}
},
});
You may notice that the variable lastUpdate is used to control that the last update in the queue is used only once.
I'm new to RxJs and need help/understanding for the following.
I have page that displays current covid cases. I have it setup to poll every 60 seconds. What I'm trying to understand is, if I subscribe to this observable via another new component, I have wait until the next iteration of 60 seconds is complete to get the data. My question is, if I want to share, is there any way to force to send the data and restart the timer?
I don't want 2 different 60 second intervals calling the API. I just want one, and the interval to restart if a new subscriber is initialized. Hope that makes sense.
this.covidCases$ = timer(1, 60000).pipe(
switchMap(() =>
this.covidService.getCovidCases().pipe(
map(data => {
return data.cases;
}),
),
),
retry(),
share(),
);
I think this should work:
const newSubscription = new Subject();
const covidCases$ = interval(60 * 1000).pipe(
takeUntil(newSubscription),
repeat(),
switchMap(() =>
this.covidService.getCovidCases().pipe(
/* ... */
),
),
takeUntil(this.stopPolling),
shareReplay(1),
src$ => defer(() => (newSubscription.next(), src$))
);
I replaced timer(1, 60 * 1000) + retry() with interval(60 * 1000).
My reasoning was that in order to restart the timer(the interval()), we must re-subscribe to it. But before re-subscribing, we should first unsubscribed from it.
So this is what these lines do:
interval(60 * 1000).pipe(
takeUntil(newSubscription),
repeat(),
/* ... */
)
We have a timer going on, until newSubscription emits. When that happens, takeUntil will emit a complete notification, then it will unsubscribe from its source(the source produced by interval in this case).
repeat will intercept that complete notification, and will re-subscribe to the source observable(source = interval().pipe(takeUntil())), meaning that the timer will restart.
shareReplay(1) makes sure that a new subscriber will receive the latest emitted value.
Then, placing src$ => defer(() => (newSubscription.next(), src$)) after shareReplay is very important. By using defer(), we are able to determine the moment when a new subscriber arrives.
If you were to put src$ => defer(() => (console.log('sub'), src$)) above shareReplay(1), you should see sub executed logged only once, after the first subscriber is created.
By putting it below shareReplay(1), you should see that message logged every time a subscriber is created.
Back to our example, when a new subscriber is registered, newSubscription will emit, meaning that the timer will be restarted, but because we're also using repeat, the complete notification won't be passed along to shareReplay, unless stopPolling emits.
StackBlitz demo.
This code creates an observable onject. I think what you should do is to add a Replaysubject instead of the Observable.
Replaysubjects gives the possibility to emit the same event when a new subscription occurs.
timer(1, 60000).pipe(
switchMap(() =>
this.covidService.getCovidCases().pipe(
tap(result => {
if (!result.page.totalElements) {
this.stopPolling.next();
}
}),
map(data => {
return data.cases;
}),
tap(results =>
results.sort(
(a, b) =>
new Date(b.covidDateTime).getTime() -
new Date(a.covidDateTime).getTime(),
),
),
),
),
retry(),
share(),
takeUntil(this.stopPolling),
).subscribe((val)=>{this.covidcases.next(val)});
This modification results in creating the timer once so when you subscribe to the subject it will emit the latest value immediately
You can write an operator that pushes the number of newly added subscriber to an given subject:
const { Subject, timer, Observable } = rxjs;
const { takeUntil, repeat, map, share } = rxjs.operators;
// Operator
function subscriberAdded (subscriberAdded$) {
let subscriberAddedCounter = 0;
return function (source$) {
return new Observable(subscriber => {
source$.subscribe(subscriber)
subscriberAddedCounter += 1;
subscriberAdded$.next(subscriberAddedCounter)
});
}
}
// Usage
const subscriberAdded$ = new Subject();
const covidCases$ = timer(1, 4000).pipe(
takeUntil(subscriberAdded$),
repeat(),
map(() => 'testValue'),
share(),
subscriberAdded(subscriberAdded$)
)
covidCases$.subscribe(v => console.info('subscribe 1: ', v));
setTimeout(() => covidCases$.subscribe(v => console.info('subscribe 2: ', v)), 5000);
subscriberAdded$.subscribe(v => console.warn('subscriber added: ', v));
<script src="https://cdnjs.cloudflare.com/ajax/libs/rxjs/6.5.3/rxjs.umd.min.js"></script>
Future possibilities:
You can update the operator easily to decrease the number in case you want to react on unsubscribers
!important
The takeUnit + repeat has already been postet by #AndreiGătej. I only provided an alternative way for receiving an event when a subscriber is added.
Running stackblitz with typescript
If the subscriberAdded operator needs some adjustements, please let me know and I will update
I'm using RxJS shareReplay() operator on an observable (courses$) to share observable stream among other two observables (beginnerCourses$ and advancedCourses$). It's working fine and single API call response is shared between both observables on success.
But, when it comes to error, these observable don't share error and error is seen to be thrown twice in the browser console. Doesn't shareReplay() operator shares error also? Is it an intended behavior?
const http$ = createHttpObservable('/api/courses');
const courses$ = http$
.pipe(
map(res => res['payload'] ),
shareReplay(),
catchError(err => {
return throwError(err);
})
);
this.beginnerCourses$ = courses$
.pipe(
map(courses => courses
.filter(course => course.category === 'BEGINNER')));
this.advancedCourses$ = courses$
.pipe(
map(courses => courses
.filter(course => course.category === 'ADVANCED')));
}
It is expected behavior when an observable throws an error, shareReplay() will try resubscribe/re-execute the source
you can try this to validate
const a=defer(()=>{
console.log('run')
return throwError(new Error('Error'))
}).pipe(shareReplay())
a.subscribe(console.log,console.error,()=>console.log('complete'))
a.subscribe(console.log,console.error,()=>console.log('complete'))
https://stackblitz.com/edit/typescript-jutfxe
If you want observable to share the error without execute again
use publishReplay(1),refCount() instead
I think it's the expected behavior, what's a bit unexpected is that you're getting 2 different errors.
shareReplay places a ReplaySubject between the data consumer(s) and the data producer. When an error notification arrives, the ReplaySubject in use will send the same error notification to all the registered subscribers:
error(err: any) {
if (this.closed) {
throw new ObjectUnsubscribedError();
}
this.hasError = true;
this.thrownError = err;
this.isStopped = true;
const { observers } = this;
const len = observers.length;
const copy = observers.slice();
for (let i = 0; i < len; i++) {
copy[i].error(err);
}
this.observers.length = 0;
}
Source
but when using shareReplay, when an error occurs the ReplaySubject in use will be replaced by another one when a new subscriber is about to subscribe. Saying that it's being replaced, this also entails re-subscribing to the source.
So I think all the subscribers should receive the same error notification as long as they are already part of the ReplaySubject's list of subscribers. Otherwise, when a new subscriber comes in, the source will be re-subscribed.
What you could do is to prevent the ReplaySubject from receiving error notifications and allow its subscribers to receive it as they are, is to use materialize and dematerialize operators:
const courses$ = http$
.pipe(
materialize(), // Everything as a `next` notification
map(res => res['payload'] ),
shareReplay(),
dematerialize() // Back to the original event
);
With this approach, if a registered subscriber receives the error notification, it will be unsubscribed, meaning that it will also be removed from the ReplaySubject's list of subscribers. But the ReplaySubject will still be there and will not be replaced on subsequent subscribers.
Also, I think this is pretty redundant:
catchError(err => throwError(err));
How to invoke a method when all the observers have unsubscribed from a subject.
Update
const alphaStore = new BehaviourSubject(0);
observer1 = alphaStore.subscribe(console.log);
observer2 = alphaStore.subscribe(console.log);
And when all of these observers unsubscribe. I want a method to be invoked. Like...
Observer1 unsubscribed
Observer2 unsubscribed
All observers left
What you describe already does the finalize() operator. Better said finalize() calls its callback when the chain disposes which means it's called when all observers unsubscribes, the chain completes or errors.
const subject = new Subject();
const shared = subject.pipe(
finalize(() => console.log('finalize')),
share(),
);
https://stackblitz.com/edit/rxjs-rebfba
When all observers unsubscribe share() unsubscribes from its source which triggers finalize().
Currently there's no way to distinguish why finalize() was invoked. See this issue https://github.com/ReactiveX/rxjs/issues/2823 and examples there on how to do it.
You can create a custom Observable, that will track the subscription count.
Heres a simple example:
let count = 0;
const tracked$ = new Observable(() => {
count++;
return ()=>{
count--;
if (count === 0) {
console.log('I am empty');
}
};
})
And then merge it with Observable that does actual work.
For simplicity sake, lets imagine its just a timer
// const tracked$ = ...
const data$ = Observable.timer(0, 5);
const result$ = data$
.merge(tracked$)
.take(5)
.subscribe(value => console.log('v:', value));
After 5 values were emitted -- it will log I am empty.
Heres a live example (with some rewrite and two subscriptions):
https://observable-playground.github.io/gist/4a7415f3528aa125fb686204041138cb
NOTE: this code uses rxjs-compat notation, which is easier to read. Above linked example uses .pipe notation, which is more common now.
Hope this helps.